U70 Preview 2: Performance Adjustments to Repeating+Dual Crossbow Styles

[l_remmie]

They come with a good argument. Shooting like a machine gun causes lag sounds plausible.
I am willing to accept significant changes to my favorite playstyle if it reduces lag and remains viable and fun.

However i am EXTREMELY skeptical that this won't come with a major DPS nerf.

 

[Reifee - Orien]

This does feel like a nerf as the Net gains are below the power loss.

However, can we REMOVE the timed delay for the start of EF ? Losing that 2-3 second delay would be a little helpful.

 

[Dvinesword]

So I have been testing it on Cormyr using the test dummies and a rough estimate is that my overall DPS is down about 10-15%. And as expected the time to develop stacks is noticeably down. It doesn't seem clunky but noticeably slower. I guess I would like to see some adjustments to the double shot increase in T5 inquisitive and maybe just a shorter windup when starting no holds barred or something like that to get the DPS back up. EDIT : or maybe increase the proc chance on diplomatic immunity to increase the stacking speed.

 

[Guntango]

I have. It comes out slightly higher overall at the moment in a pure DPS race, though like anything, actual gameplay and builds make that measure more complex - Hence why we're looking closely at today's gameplay.

Thank you.

 

[SteelStar]

In general I strongly disagree with changing game mechanics to solve performance problems. I think time would be far better spent tackling why this many requests is a problem for the server, i.e. what is actually happening under the hood and how can that be streamlined to make each request more performant. Obviously I don't know anything about your code base or network topography, but I don't think I've ever heard of another game changing its game mechanics to solve performance problems, but this is now several times you guys have done changes like this, and I think it's an alarming trend that suggests that you may not know how to actually fix the underlying problem.

In this case, the performance problem is due to the game mechanic: The rate of these ranged attacks being higher than what the servers and 20-year-old engine can handle. While there are certainly actions we're taking to improve the capacity of the game to handle things like this - 64-bit servers do help with this issue - They cannot fix the problem on their own. We could let the lag caused by this continue to affect players for much, much longer than this to get a hypothetical risky-but-broader engine change that could potentially handle those speeds, but instead we're looking at adjusting this now, with changes to offset the reduction in speed, to bring better performance fixes now instead of in the indeterminate future.

 

[Reifee - Orien]

Please confirm that you've tested dps output and it remains relatively similar after the apm adjustment.

If the Cap is 100 Doubleshot its a net loss. Its fairly easy to run at around 100 DS. They will need to remove the cap and have a third hit happen, or give an increase in the Crit profile or something to math it out.

 

[woq]

Tested on Lama (Cormyr), still looks fine. Pressing buttons isnt as fun as it was before, but this should alleviate lag a lot and the dmg loss is negligible and in some scenarios higher in single target situations

Testing done on first life repeater artificer

10/10 tbh, happy with the changes. Trash is gonna be harder, sometimes spend time waiting for an ability to resolve off cd and see it not fire off immediately, single target is similar, vast reduction in lag, overall happy woq.

 

[Rugar]

As someone who runs a lot of Xbow builds, I'm hoping this will help deal with the "disappearing shot" scenario where the animation goes off but the bolt(s) disappear into the great unknown. A few issues I hope you will consider here Steel.

(1) Great Crossbows. I know that "no one" uses them anymore since it's so much easier to get better DPS with a dual-shooter, but if you allow the speed reductions for things like Rapid Reload and Rapid Shot to affect GCs equally, you miss an opportunity to bring this type of crossbow back into parity. Leaving GCs with the live version of these feats while applying the "reduced" version to Dual- and Repeating Xbows seems to me like a net positive. I'm pretty comfortable predicting that no GC build is standing out as a source of "multiple attack" lag.

(2) Imbue builds. Since the advent of the Imbues, there has been a pretty dedicated group of people working to squeeze absolute maximum DPS out of imbue builds. As it stands now, throwers (particularly shuriken throwers) already have an advantage in the imbue builds since a single throw animation can produce multiple attacks, each with chances to have doubleshot procs. As EinarMal has noted above, if you don't balance doubleshot %s you are just going to push Xbow builds into thrower builds... not fixing the problem you note.

(3) Balancing everything out. Related to the imbue issues noted above, there is already a slight advantage to using light crossbows vs their heavy counterparts due to the attack speed if you aren't already hitting the animation cap. If you reduce the rate of fire like you are attempting here, you are going to see a bunch of people swap over to light crossbows to get back up to the max animation speed... and this is likely to defeat your goals here. To keep people using the heavy versions, I would suggest changing the damage die slightly to make up for this distance. Just looking at the most common Heavy Crossbow in epic leveling (Storm), you could change that from 3(d12)+6 to 3(2d6)+6. This will keep many people using Storm instead of swapping over to the Diplomancer (3[d8]+6) for the increase in rate of fire.

 

[Zaszgul]

There's no possible way this won't hurt dps in heroics. At cap would require testing, but there's just not enough DS in heroics to make up the difference.

Will this be accompanied by some tweaking of heroic trees, to add or move some DS to earlier levels?

 

[QuantumFX]

It's too bad we don't have boss test kobolds on the live servers to have something to compare against.

 

[Randomdude1223]

Is there still a cap of 100% on doubleshot?

If so then expanded clip is a potential big nerf if you are over 100%.

Also can we please get rid of EF windup to compensate for the pause?

To my knowledge doubleshot is not capped only doublestrike

 

[woq]

It's too bad we don't have boss test kobolds on the live servers to have something to compare against.

Yeah, thankfully I had a character with the same levels, same items and same spec as a first life repeater artificer in previous lamannia I used to test and my boss kobold timings are VERY similar to before.

 

[Ryiah]

Is there still a cap of 100% on doubleshot?

There hasn't been a cap on it since Update 28 Patch 1.

https://ddowiki.com/page/Doubleshot

 

[Teruzuki]

I want to suggest that there be a flat damage increase in the artficer battle engineer/rogue mechanic/inquisitive trees to compensate for the dps loss in heroics (maybe +3 total?). Flat damage increases are insignificant once you reach epic levels and so shouldn't be unbalancing, but I do not think there is anywhere near enough doubleshot in heroic levels (especially low heroic levels!) to compensate for the shot speed nerf. Fortunately we are not dealing with alot of damage at heroic levels, so I think this is easy to address.

One advantage of adding in a flat damage increase like this is that it can be tuned up or down if the designers find that arty/inq are over/underperforming in heroics.

Just my thought as someone who has exclusively played ranged (mainly in heroics) for over 10 years.

Thank you for addressing the lag problem! I really appreciate it!

(this is literally my first forum post)

 

[Phaaze]

because this hits lower levels harder than higher, is there a chance you can have the base attack bonus attack speed start higher but end at the same speed at cap kinda like you did with bows?

that way early levels don't feel this as much because they don't have the doubleshot to keep up in dps at those levels but still ends up where you want it at cap?

 

[StegoJR]

Is there still a cap of 100% on doubleshot?

If so then expanded clip is a potential big nerf if you are over 100%.

Also can we please get rid of EF windup to compensate for the pause?

There has not been a doubleshot cap for any ranged weapon for a while now... Of course, it was VERY hard to push a dual XBow over 200% to see it... But its easy to see the triple hits with around 140 doubleshot on archers... Also, the thrower builds thrive on crazy doubleshot stacking. Not sure where your info comes from, but only dpublestrike is capped.

 

[Phaaze]

also suggestion for the future: make some active abilities have a longer proc animation and pump their damage to compensate for lost basic attacks.

this will make actives feel more impactful and should also still bring the number of attacks/second down.

adding a cool animation to them would help too like shockwave or eldritch strike have.

 

[Habreno]

Having played a test of this on Lam (Corm). Let me preface this with the fact that I understand the intent behind the change.

This is a single-target DPS buff and a trash DPS nerf. Overall, this change plays meh at best and bad at worst. I don't need more boss DPS, I need more trash-handling ability, and these changes remove a lot of that ability. Losing shots directly removes the ability to handle trash mobs effectively, and changes such as increasing AoE sizes/damage simply is not likely to be effective enough to offset that.

Looking at ranged styles here. Bow is single-target DPS. Inquis is more hits but less damage. Repeater is further along that line. These styles want on-hit effects and quite a few of them. This change aims to push their firing rate back but provide more damage, almost seeming like you want to push them into the space bows occupy? Not sure.


If you're looking to change how procs cause lag, perhaps instead of having each shot roll for if a proc occurs, have each proc roll for the *next* occurrence, and then when that attack happens the proc happens with that attack. At that point, then roll for the new next occurrence. Rinse and repeat. This should cause less lag, since you aren't rolling to see if each proc happens on each bolt, and tracking is likely going to be less intensive than rolling.

 

[Summoner]

I had a crazy idea to keep the shooting speed and not cause lag.
I have no idea if it can be done, but what about making shot 2 and 3 of a repeater piggy back of the first one.
Like for twf.

Don't know if it makes any sense.

If you guys are going ahead and change it to what you said i hope you have done some calculations on the damage output, which should be comparable.

EDIT: Habreno Ninjaed me.

 

[pevergreen]

Couldn't reliably test my DPS but it looks and feels TERRIBLE now as a repeating xbow.

I dont need it to be lightning speed, i spent a long time as a great xbow person (and would gladly go back) but if this change is coming I'm getting out of any sort of multi shooter xbow. This feels bad, even if i dont have a measure on my dps.